Achieving traceless ablation of solid tumors without recurrence by mild photothermal-chemotherapy of triple stimuli-responsive polymer-drug conjugate nanoparticles

Article abstract of DOI:10.1039/c8tb02432d

Although photothermal therapy (PT) and photothermal-chemotherapy (PT-CT) treatments have been used to achieve complete ablation of solid tumors, they are often implemented at more than 50 °C under high intensity and using a high dose of NIR irradiation, concomitantly inducing heavy skin burning, tissue damage, and ugly scarring. Moreover, the residual tumor cells at the treated site cannot be completely eradicated, resulting in tumor recurrence and metathesis. These key obstacles have prohibited PT and PT-CT treatments from transitioning to clinical use, therefore achieving traceless ablation of solid tumors without recurrence is still a challenge for real applications. To balance hyperthermia and a high drug-loading capacity in polyprodrugs to achieve mild PT-CT, we rationally designed a novel type of intracellular pH and reduction-cleavable chlorambucil prodrug and synthesized high drug-loading polydopamine-chlorambucil conjugate nanoparticles (PDCBs). The PDCBs show good photothermal properties and demonstrate intracellular pH-, reduction-cleavable, and external near infrared (NIR)-triggered drug release profiles. Polydopamine-chlorambucil conjugate nanoparticles with 40 wt% CB (PDCB₄₀) and mild NIR irradiation could facilitate cellular internalization and subcellular trafficking, generating an excellent and synergistic antitumor effect in vitro. Pharmacokinetics and small animal fluorescent and photoacoustic imaging demonstrate that PDCB₄₀ has a 3.6-fold longer blood circulation time compared to free CB and attained selective tumor accumulation, simultaneously inducing a 4.1-fold stronger photoacoustic signal than the control. By using one intravenous injection of PDCB₄₀ and a single dose of mild NIR irradiation, this simple and mild PT-CT treatment achieved a non-discerned tumor on the sixth day, and traceless and complete ablation of a solid MCF-7 tumor without recurrence within 50 days, opening up a new avenue for precise cancer therapy with the potential for real applications.

Full text of DOI:10.1039/c8tb02432d

Journal of
Materials Chemistry B
View Article Online
View Journal | View Issue
PAPER
Achieving traceless ablation of solid tumors
without recurrence by mild photothermal-
chemotherapy of triple stimuli-responsive
polymer–drug conjugate nanoparticles†
Cite this: J. Mater. Chem. B, 2019,
, 415
7
a
a
a
b
ab
Chang Du, Yue Ding, Jiwen Qian, Rong Zhang and Chang-Ming Dong
*
Although photothermal therapy (PT) and photothermal-chemotherapy (PT-CT) treatments have been
used to achieve complete ablation of solid tumors, they are often implemented at more than 50 1C
under high intensity and using a high dose of NIR irradiation, concomitantly inducing heavy skin burning,
tissue damage, and ugly scarring. Moreover, the residual tumor cells at the treated site cannot be
completely eradicated, resulting in tumor recurrence and metathesis. These key obstacles have
prohibited PT and PT-CT treatments from transitioning to clinical use, therefore achieving traceless
ablation of solid tumors without recurrence is still a challenge for real applications. To balance
hyperthermia and a high drug-loading capacity in polyprodrugs to achieve mild PT-CT, we rationally
designed a novel type of intracellular pH and reduction-cleavable chlorambucil prodrug and synthesized
high drug-loading polydopamine-chlorambucil conjugate nanoparticles (PDCBs). The PDCBs show good
photothermal properties and demonstrate intracellular pH-, reduction-cleavable, and external near
infrared (NIR)-triggered drug release profiles. Polydopamine-chlorambucil conjugate nanoparticles with
4
0 wt% CB (PDCB40) and mild NIR irradiation could facilitate cellular internalization and subcellular
trafficking, generating an excellent and synergistic antitumor effect in vitro. Pharmacokinetics and small
animal fluorescent and photoacoustic imaging demonstrate that PDCB40 has a 3.6-fold longer blood
circulation time compared to free CB and attained selective tumor accumulation, simultaneously
inducing a 4.1-fold stronger photoacoustic signal than the control. By using one intravenous injection of
PDCB40 and a single dose of mild NIR irradiation, this simple and mild PT-CT treatment achieved a non-
discerned tumor on the sixth day, and traceless and complete ablation of a solid MCF-7 tumor without
recurrence within 50 days, opening up a new avenue for precise cancer therapy with the potential for
real applications.
Received 14th September 2018,
Accepted 5th December 2018
DOI: 10.1039/c8tb02432d
rsc.li/materials-b
radiotherapy, chemotherapy (CT), and their cocktail therapies,
this NIR-mediated PT treatment presents several advantages,
1
. Introduction
The near-infrared light (NIR, 650–950 nm) absorbing nano- such as: (1) achieving spatiotemporal manipulation, pinpoint
particles including inorganic carbon and gold nanostructures, tumor ablation, and a passive targeting effect to the tumor
transition-metal dichalcogenides, semiconducting polymers, via the enhanced permeation and retention effect (EPR); (2) by
melanin and polydopamine, to name just a few, can efficiently thermally expanding the tumor blood vessels and enhancing
transduce laser irradiation to hyperthermia and achieve cancer the membrane permeability, the hyperthermia effect facilitates
1
–12
photothermal therapy (PT) in a minimally invasive manner.
the extravasation, accumulation and penetration of nanoparticles
Compared with various cancer treatments including surgery, in the tumor, and subsequent cellular internalization; and (3) this
hyperthermia might induce antitumor immunity, improve the
a
efficacy of CT, and generate a synergistic antitumor effect in the
School of Chemistry and Chemical Engineering, Shanghai Key Laboratory of
Electrical Insulation and Thermal Aging, Shanghai Jiao Tong University,
Shanghai 200240, P. R. China. E-mail: cmdong@sjtu.edu.cn
1
3–16
cocktail therapies.
Although most of the PT and PT-CT
treatments have achieved complete ablation of solid tumors, they
are often implemented at more than 50 1C under high intensity
and a high dose of NIR irradiation, concomitantly inducing heavy
and ugly scarring, skin burning, and damage to the surrounding
b
Joint Research Center for Precision Medicine, Shanghai Jiao Tong University
Affiliated Sixth People’s Hospital South Campus, Shanghai Fengxian Central
Hospital, Shanghai 201400, P. R. China
†
Electronic supplementary information (ESI) available: Scheme S1, Table S1, S2
1
7–20
and Fig. S1–S15. See DOI: 10.1039/c8tb02432d
normal tissue.
Moreover, the residual tumor cells in the
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 415

View Article Online
Paper
Journal of Materials Chemistry B
treated site cannot be completely eradicated, resulting in tumor that have a high drug-loading capacity and good NIR-absorbing
recurrence and metathesis. These key obstacles prohibit PT and attributes (Fig. 1), maximizing the intracellular physiological
its cocktail treatments from transitioning to the clinic, and stimuli-triggered CT efficacy under mild photothermia (i.e.,
achieving traceless ablation of solid tumors without recurrence endolysosome pH 5.0 and 10 mM DTT reduction to mimic
1
6,27,43–46
using mild PT (i.e., mild photothermia under mild intensity the cytosolic glutathione gradient).
Using one intra-
and a dose of NIR irradiation) is urgently needed to enable real venous injection and a single dose of mild NIR irradiation
ꢀ
2
application of this widely studied PT technology.
(808 nm, 1 W cm , 10 min), the polyprodrug polydopamine-
In the case of very mild PT (i.e., weak PT under a lower chlorambucil conjugate nanoparticles with a 40 wt% CB (PDCB40
)
irradiation intensity and lower dose), a few examples from the formulation generated a superior and synergistic antitumor
published research have made progress in overcoming the efficacy and achieved complete and traceless ablation of a solid
above-mentioned obstacles. Liu et al. reported gambogic acid MCF-7 tumor that did not recur within 50 days, providing a
and indocyanine co-loaded coordination polymer nanofibers simple, practical and mild PT-CT treatment for use in precision
induced efficient tumor cell apoptosis at approximately 43 1C cancer therapy.
ꢀ2
after 20 min of 808 nm irradiation at 0.3 W cm , during which
two out of five tumors were completely eradicated in a 14 day
period with the assistance of gambogic acid to inhibit heat- 2. Experimental section
1
9
shock protein production. With 21 days of consecutive NIR
2
.1 Materials
ꢀ
2
irradiation (10 min, 850 nm, 0.4 W cm ), a very mild tempera-
ture of B44 1C was used to achieve elimination of a 4T1 tumor
using a pH-sensitive hemolytic polymer coated polydopamine
Acetone (99.5%), dichloromethane (99.5%), diethyl ether (99.5%),
and methanol (99.5%) were distilled before use, respectively.
Chloroambucil (CB, 98%, J&K), dicyclohexylcarbodiimide (DCC,
(
2
PDA), the tumor disappeared on the 15th day, within the
1 day therapeutic period. Therefore, developing mild PT to
2
0
99.2%, Aldrich), 4-dimethylaminopyridine (DMAP, 99%, Adamas),
0
3,4-dihydroxy-benzenepropanoic acid (98%, Aldrich), 2,2 -dithio-
achieve traceless and complete eradication of solid tumors
without recurrence is still challenging, especially to provide
simple method for use in the clinic (i.e., simple treatment
protocol including one intravenous injection with a single NIR
irradiation dose).
diethanol (Alfa, 90%), D,L-dithiothreitol (DTT, 99%, J&K), dopamine
hydrochloride (98%, Aldrich), phosphorus trichloride (99%, Aldrich),
and trometamol (tris, Z99.9%, Aldrich) were used as received.
Trifluoroacetic acid (Z99%) was purchased from Shanghai
Sinopharm Corp. Dulbecco’s modified eagle medium (DMEM,
PAA laboratories), fetal bovine serum (FBS, PAA laboratories),
methylthiazolyldiphenyl-tetrazolium bromide (MTT, ultrapure,
Aldrich), LysoTracker (ultrapure, Yeasen), and propidium iodide
In contrast to non-biodegradability and potentially long-
term toxicity of inorganic nanomaterials, organic indocyanine
green, porphysomes, natural melanin, and biomimetic PDA
have attracted significant attention in the field of cancer PT
2
1–34 (PI, ultrapure, Yeasen) were used as received. L929 (a mouse
fibroblastic cell line), HeLa (a human uterine cervix carcinoma
owing to their good biodegradability and biocompatibility.
Owing to their simple self-polymerization (i.e., one-pot syn-
thesis of nanoparticles in a mild alkaline solution), good
biocompatibility and biodegradability, and their biomimicry
of melanin (i.e., the hierarchical structure and NIR-absorbing
function), the PDA and derived materials hold great promise
in cancer PT, PT-CT, and theranostics. On the other hand,
cell line), and MCF-7 (a human breast carcinoma cell line)
were received from Shanghai Institute of Biochemistry and Cell
Biology.
2.2 Methods
polymer–drug conjugates provide distinct attributes including Fourier transform infrared (FT-IR) spectroscopy was recorded
variable drug-loading capacity, no leaking of the drug to be on a Perkin Elmer Spectrum 100 spectrometer at room tem-
1
released, minimized side effects and an enhanced CT efficacy perature. H nuclear magnetic resonance (NMR) (400 MHz) and
3
5–42
13
compared to the physically drug-loaded materials.
Several
C NMR (100 MHz) spectra were recorded on a Varian Mercury-
polymer–drug conjugates (or polyprodrugs) are being trans- 400 spectrometer at room temperature using tetramethylsilane
lated into the clinic, however, polyprodrugs with an intrinsic as an internal standard. Time-of-flight mass spectrometry
NIR-absorbing capacity and photo-mediated imaging functions (TOF-MS) was performed on a Waters Acquity UPLC/Premier
(
e.g., photoacoustic imaging) are rarely constructed for use in QTOF MS Premier. X-ray photoelectron spectroscopy (XPS) was
4
3
precision cancer PT and PT-CT treatment. To address the two conducted on a VG ESCALAB MKII spectrometer and obtained
key dilemmas in cancer PT and PT-CT treatments, that is, the by using XPS PEAK software (Version 4.1) to deconvolute the
first is that the high intensity and high dose of NIR irradiation spectra. The mean hydrodynamic diameter and polydispersity
induce heavy scarring, skin burning, and tissue damage, and index (PDI) of the nanoparticles were determined using dynamic
the second is that the residual tumor cells in the treated site light scattering (DLS), during which the samples were measured
can result in tumor recurrence and metathesis, we put forward five times on a Malvern ZS90 instrument at 25 1C. Transmission
an innovative strategy of leveraging the hyperthermia and the electron microscopy (TEM) was performed on a JEM-2010 instru-
high drug-loading capacity of polyprodrugs to achieve mild ment at a 200 kV accelerating voltage, during which the samples
PT-CT. We rationally designed a novel dual stimuli-sensitive were dropped onto 300 mesh Formvar-carbon film-coated
polydopamine-chlorambucil conjugate nanoparticles (PDCBs) copper grids. The UV-Vis-NIR spectroscopy was recorded on a
4
16 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 1 Synthesis of the prodrug dopamine-chloroambucil (DA-CB) and the PDCBs conjugate nanoparticles (A). The cellular uptake of the PDCBs,
intracellular pH and GSH cleavable CB release, and the synergistic PT-CT upon mild NIR irradiation (B). Illustration showing mild PT-CT treatment with
one injection and one dose of NIR irradiation and both PT and PA imaging in vivo (C).
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 417

View Article Online
Paper
Journal of Materials Chemistry B
Perkin-Elmer Lambda 750S spectrometer at room temperature. (0.5 mmol, 179.0 mg) in dichloromethane (1 mL). The reaction
The fluorescent spectroscopy was recorded on a Perkin-Elmer mixture was vigorously stirred for another 48 h under a nitrogen
LS-50B spectrometer at room temperature. All of the 808 nm atmosphere at room temperature. Dichloromethane was removed
NIR irradiation was carried out by using a continuous wave by rotary evaporation and the residues were purified using silica
diode laser (Shanghai SFOLT Corp., FC-960-6000-MM) with column chromatography (dichloromethane) to give the product
1
tunable power (0–1650 mW), during which the spot size was (85% yield). H NMR (400 MHz, CD
3
OD), d (ppm) = 1.60 (s, 6H,
tuned using a fiber collimator connected by a fiber optic patch (CH
cable (FC/PC/200UM/1M). 2H, CCH
CCH CH -Ar), 2.87 (m, 6H, CH
.60–3.68 (m, 4H, N(CH CH Cl) ), 4.32 (m, 4H, CH
.60 (m, 5H, Ar), 7.07 (m, 2H, Ar). C NMR (100 MHz, CD OD),
3
)
2
), 1.90 (m, 2H, CCH
CH CH -Ar), 2.57 (m, 4H, CCH
SSCH and CCH
CH
2 2 2
CH CH -Ar), 2.38 (dd, J = 7.6, 7.2 Hz,
2
2
2
2
CH
2
CH
2
-Ar and
CH -Ar),
CH ),
2
2
2
2
2
2
2
.3 Synthesis of the chloroambucil prodrug of
dopamine-chloroambucil
The dual stimuli-sensitive prodrug dopamine-chloroambucil d (ppm) = 173.55 (COO), 172.81 (OOC), 147.70 (Ar), 145.88 (Ar),
DA-CB) was synthesized using the following four steps. Firstly, 144.48 (Ar), 133.27 (Ar), 130.94 (Ar), 129.80 (Ar), 120.52 [C(CH )],
3
6
2
2
2
2
2
SSCH
2
2
13
3
(
3
3
,4-dihydroxybenzenepropanoic acid (DBA, 1.0 g, 5.49 mmol) 117.69 (Ar), 112.57 (Ar), 111.83 (Ar), 109.11 (Ar), 107.72 (Ar), 62.30
was dissolved in dry acetone (20 mL) at 0 1C. Phosphorus (COOCH ) 61.02 (CH COO), 53.81 (CH Cl), 53.51 (CH Cl), 40.91
trichloride (383 mL, 4.4 mmol) was then slowly added dropwise (NCH ), 40.45 (NCH ), 37.43 (SCH ), 37.42 (CH S), 36.49 (ArCH2-
into the above solution and the reaction mixture was vigorously CH CO), 34.38 (ArCH CH CH CO), 33.21 (ArCH CH CH CO),
stirred for 6 h at 0 1C. The acetone was removed by rotary 29.98 (ArCH CH CO), 26.92 (CH CH CH ), 25.82 [C(CH )]. FT-IR
evaporation and the residues were dissolved in diethyl ether (KBr, cm ): 2981 (nC–H), 1732 (nCOO), 1613, 1518 (nphenyl), 1076
10 mL). The organic layer was extracted using water three times (nC–O–C), 802 (nC–Cl).
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
ꢀ
1
(
and then concentrated to give 2,2-dimethyl-1,3-benzodioxole-
Finally, the above-obtained product (0.5 mmol, 322.0 mg) was
4
7 1
5
-propanoic acid (65% yield).
d (ppm) = 1.60 (s, 6H, (CH
CCH CH COOH), 2.71 (dd, J = 7.6, 7.6 Hz, 2H, CCH
COOH), 6.70–6.60 (m, 3H, Ar). C NMR (100 MHz, DMSO-d
d (ppm) = 174.28 (CQO), 147.26 (Ar), 145.53 (Ar), 134.50 (Ar), 2.33 (dd, J = 7.6, 7.2 Hz, 2H, CCH
20.91 (Ar), 117.96 C(CH ), 108.95 (Ar), 108.23 (Ar), 36.05 CCH CH CH -Ar and CCH CH -Ar), 2.87 (m, 6H, CH SSCH and
H NMR (400 MHz, CDCl
) ), 2.47 (dd, J = 7.6, 4.8 Hz, 2H, added dropwise and stirred for 24 h at room temperature. The
3 2
3
), dissolved in dichloromethane (5 mL) and then TFA (5 mL) was
1
2
2
2
CH
2
-
target product of DA-CB was quantitatively obtained. H NMR
), (400 MHz, CD OD), d (ppm) = 1.90 (m, 2H, CCH CH CH -Ar),
CH CH -Ar), 2.57 (m, 4H,
1
3
6
3
2
2
2
2
2
2
1
3
2
2
2
2
2
2
2
ꢀ
1
(Ar-CH ), 30.59 (CH CO), 25.99 C(CH ). FT-IR (KBr, cm ): CCH CH -Ar), 3.60–3.68 (m, 4H, N(CH CH Cl) ), 4.32 (m, 4H,
2 2 3 2 2 2 2 2
1
3
2981 (nC–H), 1707 (nCOOH), 1609, 1498 (nphenyl), 1122 (nC–O–C).
CH CH SSCH CH ), 6.60 (m, 5H, Ar), 7.07 (m, 2H, Ar). C NMR
2
2
2
2
3
Secondly, 2,2-dimethyl-1,3-benzodioxole-5-propanoic acid (100 MHz, CDCl ), d (ppm) = 174.88 (C1), 174.49 (C2), 144.00
(
(
1 mmol, 222.0 mg), DMAP (1 mmol, 122.2 mg), and DCC (C3), 142.47 (C4), 139.97 (C5), 136.74 (C6), 133.18 (C7), 130.53
1 mmol, 206.0 mg) were dissolved in dichloromethane (5 mL) (C8, C9), 120.94 (C10), 116.12 (C11, C12), 115.39 (C13,C14), 62.63
and stirred for 4 h at room temperature under nitrogen atmo- (C15, C16), 56.11 (C17, C18), 39.06 (C19, C20), 37.39 (C21), 36.23
sphere. Then the above solution was added dropwise into (C22), 34.36 (C23), 33.70 (C24), 30.44 (C25), 29.13 (C26), 26.56
0
ꢀ1
2
,2 -dithiodiethanol (3 mmol, 396.5 mg) in dichloromethane (C27). FT-IR (KBr, cm ): 3409 (broad, nOH), 2981 (nC–H), 1732
(
1 mL) and the reaction mixture was vigorously stirred for (nCOO), 1615, 1519 (nphenyl), 806 (nC–Cl). TOF-MS: calculated,
+
another 48 h under a nitrogen atmosphere at room temperature. 604.1361 [M + H]; found, 604.1376 [M + H] .
Dichloromethane was removed by rotary evaporation and the
2.4 Synthesis of the PDCBs conjugate nanoparticles
residues were purified using silica column chromatography
dichloromethane: methanol, 50: 1) to give the intermediate The PDCBs conjugate nanoparticles (the subscript denotes the
(
1
product (70% yield). H NMR (400 MHz, CD3OD-d
1
4
), d (ppm) = weight percentage of the CB) and the control PDA ones were
CH ), 2.83 synthesized according to the method described in our previous
.60 (s, 6H, (CH ), 2.60 (dd, J = 7.2, 7.6 Hz, 2H, ArCH
)
3 2
2
2
4
3
(m, 4H, CH S–SCH ), 2.90 (dd, J = 6.4, 6.4 Hz, 2H, ArCH ), 3.77 publication, respectively. As a typical example, both DA-CB
2
2
2
(
2
dd, J = 6.4, 6.4 Hz, 2H, S–SCH CH OH), 4.31 (dd, J = 6.4, 6.4 Hz, (53.7 mg) and DA hydrochloride (12.5 mg) were pre-dissolved in
2 2
1
3
H, S–SCH CH O), 6.60 (m, 3H, Ar). C NMR (100 MHz, deionized water (1 mL) and were then added into an aqueous
2 2
CDCl
3
), d (ppm) = 173.02 (CQO), 148.37 (Ar), 148.16 (Ar), solution (20 mL) of tris (410.0 mg) at 30 1C, and then stirred
)], 108.65 (Ar), 107.70 under air for 24 h. The resultant suspension was dialyzed
OH), 40.88 (SCH CH OH), (molecular weight cutoff: 3500) against deionized water or
), 30.18 (ArCH CH ), 24.56 PBS (10 mM, pH 7.4) for 2 days to give the nanoparticles in
)]. FT-IR (KBr, cm ): 2981 (nC–H), 1736 (nCOO), 1608, solution and then stored at ꢀ4 1C for further use.
498 (nphenyl), 1076 (nC–O–C). In addition, to prepare the DOX-loaded PDCB for small
Thirdly, chloroambucil (CB, 0.5 mmol, 152.0 mg), DMAP animal fluorescent imaging, 2 mg PDCB40 was dispersed in
1
(
3
[
33.93 (Ar), 120.51 (Ar), 117.22 [C(CH
Ar), 62.00 (CO–C–CH ), 60.00 (CH
7.07 (SCH CH O), 35.97 (ArCH CH
C(CH
3
2
2
2
2
2
2
2
2
2
2
ꢀ1
3
1
40
ꢀ1
(0.5 mmol, 61.1 mg), and DCC (0.5 mmol, 103.0 mg) was dis- PBS, followed by the addition of 3 mL of DOX in PBS (1 mg mL )
solved in dichloromethane (5 mL) and stirred for 4 h at room and this was stirred at room temperature for 24 h. Then DOX-
temperature under a nitrogen atmosphere, and then the above loaded PDCB40 was harvested by immediate centrifugation (10 min,
solution was added dropwise into the above-obtained intermediate 15 000 rpm) and then washed using PBS three times. The DOX
4
18 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
content was determined at 480 nm absorbance using UV-Vis the basis of the CB concentration; and the IC50 values for the
spectroscopy and repeated six times.
single PT and the combination PT were calculated on the basis
of the concentration of pure PDA or PDA in the PDCBs. The
combination index (CI) was calculated using the following
2.5 NIR-mediated photothermal property of PDCBs
Generally, 200 mL of the PDCBs solution with different concen- equation: CI = [IC50 (combination CT)/IC (CT)] + [IC50 (combi-
50
4
8–51
trations was added in a 96-well plate and then irradiated using nation PT)/IC50 (PT)].
All cell viabilities were tested using
ꢀ
2
the NIR laser (808 nm, 1, 2, 4 W cm , 10 min), during which six replicates.
the solution temperature and the repeated heating–cooling
cycles (i.e., 10 min NIR irradiation and 10 min naturally 2.8 Cell internalization
cooling) were recorded using a digital thermometer every 30 s,
respectively.
Using MCF-7 as the model cell line, the cell internalization
process of the free CB or PDCB40 was studied using flow cytometry
(
CB: excitation at 345 nm and emission at 435 nm, BD Accuri C6)
2
.6 In vitro chloroambucil release
and inverted fluorescence microscopy (LeicaDMI6000 B). Briefly,
For the DTT, pH, and the DTT + pH triggered release, the
PDCBs in PBS (1 mL, 10 mM, pH 7.4) with/without 10 mM DTT
were transferred into a dialysis bag (MWCO = 3500 Da), and
then incubated in 10 mL of PBS at different pH values (i.e.,
pH 5.0 and 7.4) at 37 1C. For the NIR-triggered release, the
samples were vertically irradiated using the NIR laser (808 nm,
5
MCF-7 was incubated in a 6-well plate (5.0 ꢁ 10 cells per well)
for 24 h. The same drug concentration of free CB or PDCB40
ꢀ1
(
10 mg mL CB equivalent) was added to each well for a pre-
determined time at 37 1C with or without the NIR irradiation.
The cells were rinsed with PBS, disposed with trypsin, and then
4
the data collected for the 1.0 ꢁ 10 gated events were analyzed
ꢀ
2
1
1
W cm , 10 min) at 0.5, 1, 2, 4, 6, 8, 10, 12 h and then put into
using FlowJo software. After similar treatment, MCF-7 was fixed
with 4% formaldehyde for 30 min at room temperature, washed
with PBS, stained using PI for 15 min, and then observed and
analyzed using ImageJ software.
0 mL of PBS with a constant shaking rate of 150 rpm at 37 1C.
Then, 10 mL of dialysis solution was taken out at predeter-
mined time intervals and the same volume of fresh PBS was
added back. The released CB was measured by the absorbance
at 258 nm using a UV-Vis spectrophotometer. All release
experiments were carried out in duplicate and each sample
was measured three times. In addition, the molecular structure
of the released drug from the nanoparticles was confirmed
using TOF-MS.
2
.9 Intracellular trafficking of PDCBs
Intracellular trafficking and distribution of PDCB40 was observed
using confocal laser scanning microscopy (CLSM). Briefly, MCF-7
5
was incubated in 6-well plates (5.0 ꢁ 10 cells per well) for 24 h,
replaced with fresh culture medium, then followed by addition of
ꢀ1
PDCB40 at 10 mg mL CB equivalent. After 2 h incubation, the
2
.7 In vitro PT, CT, and PT-CT
medium was removed and the cells were irradiated using the
The L929, HeLa, and MCF-7 cell lines were cultured at 37 1C in
DMEM supplemented with 10% FBS, 100 IU per mL penicillin,
and 100 mg per mL streptomycin under a humidified atmo-
ꢀ
2
NIR laser (808 nm, 1 W cm , 10 min). The cells were rinsed
with PBS three times and both cell nuclei and lysosomes were
counterstained with PI (red) and LysoTracker (green), respec-
tively. Finally, the cells were washed with PBS three times and
observed using CLSM. The data were analyzed using ImageJ
software. A control, without irradiation, was also carried out
under same conditions.
sphere of 5% CO . L929 was only used to evaluate the cytotoxi-
2
city of the PDA and the latter two carcinomas were used to
measure the cytotoxicity and antitumor activity. Briefly, the
4
cells were seeded into 96-well plates (10 cells per well) in 200 mL
of culture medium and incubated overnight at 37 1C. 200 mL of
PDA or PDCBs with gradient concentrations were added and
then incubated for 48 h. After twice washing with fresh culture
2.10 Animals
ꢀ
1
Five week-old male Balb/c nude mice (B20 g) and Sprague-
Dawley (SD) rats (B200 g) were purchased from the Chinese
Academy of Sciences (Shanghai, China). The animal experi-
ments were performed in accordance with the guidelines for
the care and use of laboratory animals and approved by the
Animal Ethics Committee of Shanghai Jiao Tong University.
medium and PBS, 20 mL of the 5 mg mL MTT stock solution in
PBS was added to each well and further incubated for 4 h to
allow full formation of formazan. Finally, the blue formazan was
dissolved in 100 mL DMSO for 20 min and the absorbance at
4
90 nm was measured using a Microplate Reader (Elx800, BioTek
Company).
To evaluate the PT and PT-CT, nanoparticle solutions with
gradient concentrations were added separately into a 96-well
plate and incubated for 4 h at 37 1C prior to the NIR irradiation The SD rats were randomly divided into two groups (n = 4) and
2.11 Pharmacokinetics
ꢀ2
ꢀ1
(
808 nm, 1 W cm , 10 min). The cells were further incubated each mouse was injected via the tail vein with 8 mg kg free CB
for 12 h for PT and 48 h for PT-CT, respectively, and then the or a PDCB40 equivalent. Blood samples (0.3 mL) were obtained
MTT assay was used to determine the cell viability. The half via eye puncture at selected time intervals (i.e., 0.25, 0.5, 1, 2,
maximal inhibitory concentration (IC50) was calculated using 4, 6, 8, 12, 14 and 24 h). Plasma samples were harvested by
GraphPad Prism 6 software from eight samples. As for the immediate centrifugation (10 min, 3000 rpm), frozen at ꢀ20 1C,
single CT and the combination CT, the IC50 was calculated on and then the CB in the blood serum was analyzed using
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 419

View Article Online
Paper
Journal of Materials Chemistry B
fluorescence spectroscopy. The pharmacokinetic parameters average ꢂ standard error. When the tumor volume of the PBS
3
were calculated using OriginPro 8.
group reached approximately 1500 mm on day 32, one mouse
from each group was sacrificed for comparative studies. Finally,
the tumors, as well as the major organs (heart, liver, spleens,
2.12 In vivo multiple imaging and biodistribution
MCF-7 tumor-bearing nude mice with a tumor volume of lung and kidneys), were dissected, washed with saline, weighed,
3
approximately 300 mm were intravenously injected via the tail photographed, and fixed with 4% formaldehyde for histology,
ꢀ1
vein with 200 mL of DOX-loaded PDCB40 (1 mg mL ) or free TUNEL, and PCNA assays. All of the mice were euthanized if the
ꢀ
1
3
DOX (5.3 mg kg ). In vivo fluorescent imaging was monitored tumor volume was larger than 1500 mm or if the body weight
at 1, 2, 4, 6, 8, 10, 12, 14 and 24 h post-injection using a Kodak loss was greater than 15%. Statistical analyses were performed
multimode imaging system (DOX: excitation wavelength = using GraphPad Prism software 5.0 and data were given as
5
30 nm, emission wavelength = 600 nm). For the ex vivo mean ꢂ S.D.
distribution, after the mice were sacrificed by cervical vertebra
dislocation at 2, 6 and 10 h post-injection (n = 3), all tumor and 2.14 Statistical analysis
major organs were carefully collected, washed with saline, and
The two-way analysis of variance and the t-test were used to
then imaged using a Kodak multimode imaging system. Tissue
samples were rinsed in saline, blotted using a paper towel,
weighed and stored at ꢀ80 1C. After being homogenized in
determine the statistical significance. A probability (P) value o
0
.01 was considered significant, and P o 0.001 was highly signi-
ficant. All computations were made by employing Microsoft
Excel 2007.
0
.5 mL of 1% Triton X-100, 1.2 mL of the extraction solution
(
HCl-IPA) was added and then incubated at ꢀ20 1C overnight.
After vortexing and centrifugation at 15 000 ꢁ g for 15 min, the
DOX amount measured using fluorescence spectroscopy is given
as the percentage injected dose per gram of tissue (% ID/g).
For the in vivo hyperthermia and imaging, the mice (n = 4)
3
. Results and discussion
3
.1 Synthesis of dual stimuli-sensitive polydopamine-
chloroambucil conjugate nanoparticles (PDCBs)
PDA or the control PBS, and then irradiated using the NIR laser As the disulfide bond is stable in physiological media and can
ꢀ1
were intravenously injected with 200 mL (2 mg mL ) of PDCB40
,
ꢀ2
(
808 nm, 1 W cm , 10 min) at 10 h post-injection. The be cleaved by the GSH gradient inside the cell environment, we
thermographic image and the temperature of the tumor site selected it to connect 3,4-dihydroxybenzenepropanoic acid with
were recorded using an infrared thermal camera (AXT100, Ann an anticancer drug CB to afford a dopamine-derived prodrug
4
4–46
0
Arbor Sensor Systems).
DA-CB.
Meanwhile, 2,2 -dithiodiethanol containing a
ꢀ
1
After PDCB40 (200 mL, 1 mg mL ) or the control PDA was disulfide bond bridge was directly linked to the carboxy group
intravenously injected into the mice (n = 3), photoacoustic (PA) of CB via an ester bond, which endowed the prodrug DA-CB
imaging was acquired at 1, 2, 4, 6, 8, 10, 12, 14 and 24 h post- with a subcellular (e.g., endolysosome, pH = 4.5–6.5) acid-labile
3
8,45
injection using a commercial Endra Nexus 128 PA tomography linker of b-thioethyl methanoate.
Thus, the prodrug DA-CB
system (Endra Inc., Ann Arbor, Michigan) equipped with a was expected to be cleaved by physiologically related dual
tunable nanosecond pulsed laser (808 nm, 7 ns pulse, 7 mJ per stimuli triggers (i.e., intracellular pH and GSH), which was
pulse, 2 Hz pulse repetition frequency) and a 128 unfocused rationally designed and synthesized using four steps and its
ultrasound transducer with a 5 MHz center frequency. The PBS molecular structure was fully characterized by means of FT-IR,
1
13
alone was used to eliminate the background signal for obtain-
ing the PA signal of the PDA and/or PDCBs.
H NMR, C NMR, and TOF-MS spectroscopy (Fig. 2A and B
1
and Fig. S1, ESI†). The H NMR spectroscopy for DA-CB
shows proton signals at 4.32 ppm (m, 4H, –COO–CH
S–CH CH –OOC–) which can be assigned to the methylene
The MCF-7 tumor-bearing mice with a tumor volume of linked to the carboxylic ester and the disulfide bond and those
2 2
CH –S–
2.13 In vivo antitumor activity
2
2
3
6
0–80 mm were randomly divided into seven groups (n = 4) at 3.60–3.68 ppm (m, 4H, –N(CH CH Cl) ) can be assigned to
2 2 2
1
3
for antitumor studies. All of the mice were intravenously CB. The C NMR spectroscopy presents the carbon signal of
injected once at a time point of 0 days with 8 mg kg CB, methylene at 62.63 ppm (–COO–CH CH –S–S–CH CH –OOC–),
ꢀ
1
2
2
2
2
ꢀ
1
nanoparticles (PDA or PDCB40, 200 mL, 2 mg mL ) or PBS. For those at 174.88 ppm and 174.49 ppm represent the two
the PDA + NIR and PDCB40 + NIR groups, the tumor sites were carboxylic esters, and those at 56.11 ppm [–N(CH CH Cl)
] represent the CB. Moreover,
0 h. The tumor size and body weight of each mouse were the TOF-MS confirms the prodrug DA-CB has an observed value
2
2
2
]
ꢀ2
2 2 2
irradiated using the NIR laser (808 nm, 1 W cm , 10 min) at and 39.06 ppm [–N(CH CH Cl)
1
+
measured by using a caliper or a balance every two days. Tumor of 604.1376 for [M + H] , which matches well with the theore-
volume (V) was calculated according to the following equation: tical value of 604.1361.
3
V (mm ) = 1/2 ꢁ length (mm) ꢁ width (mm) ꢁ width (mm). The
The oxidative self-polymerization and copolymerization of the
tumor inhibitory rates (TIR) of various treatments were calcu- dopamine and its derivatives allow the direct formation of the
lated using the equation: TIR (%) = 100 ꢁ (mean tumor volume negatively charged PDA and the related nanoparticles in weakly
2
1–24,43
of the PBS group ꢀ mean tumor volume of others)/(mean basic water (e.g., tris and NH
3 2
ꢃH O).
Thereafter, the PDCBs
tumor volume of the PBS group). Data are represented as the were prepared using the precipitation copolymerization of
4
20 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
1
13
Fig. 2 H NMR (A) and C NMR (B) spectra for the DA-CB prodrug, FT-IR (C) of PDCBs and PDA, and XPS (D) for PDCB40
.
Table 1 Characterization of the PDCBs and PDA nanoparticles
a
b
b
c
ꢀ1
c
ꢀ1
d
ꢀ1
e
Sample Diameter (nm) TEM morphology Zeta potential (mV) IC50 (PT) (mg mL
)
IC50 (CT) (mg mL
)
IC50 (PT-CT) (mg mL
)
CI
PDA
75 ꢂ 3
Sphere
Sphere
Sphere
ꢀ33.1
ꢀ32.1
ꢀ30.2
53.76 (HeLa)
9.34 (MCF-7)
53.76 (HeLa)
9.34 (MCF-7)
53.76 (HeLa)
9.34 (MCF-7)
—
—
—
6
PDCB30 126 ꢂ 5
PDCB40 154 ꢂ 1
14.15
9.92
14.03
9.72
8.42 + 3.61
8.45 + 3.62
5.54 + 3.69
5.23 + 3.49
0.41
0.49
0.37
0.43
6
6
a
b
The subscript denotes the weight percentage of CB within the PDCBs. The mean diameter and the zeta potential of the nanoparticles in PBS are
c
determined using DLS. IC50 is calculated from Fig. 6 and Fig. S10 (ESI) by using GraphPad Prism 6 software; note that the IC50 (PT) for a single PT
of PDCBs has to be substituted by PDA as an equivalent otherwise it would be disturbed by the PDCBs release induced CT. For the combination
PT-CT, the former value denotes the combination PT and the latter one is the combination CT. CI denotes the combination index of PT-CT, and
CI = [IC50 (combination CT)/IC50 (CT)] + [IC50 (combination PT)/IC50 (PT)].
d
e
48–51
DA-CB with DA in Tris aqueous solution at 30 1C for 24 h and the
The as-synthesized PDCBs were fully characterized by means
detailed results are summarized in Table 1 and Table S1 (ESI†). The of FT-IR, XPS, DLS, and TEM. Compared to the pure PDA
PDCBs have a higher NIR-mediated temperature-elevating ability counterpart, the FT-IR spectra for the PDCBs show a character-
ꢀ
1
(
DT), which would induce a heterogenous and quick photothermal istic vibration peak at 1732 cm for the ester carbonyl (–COO–)
ꢀ
1
heating effect, resulting in heavy skin and tissue damage during the and one fingerprint signal at 804 cm for CB (C–Cl), along with
17–20
ꢀ1
PT process.
Moreover, the PDA content within the PDCBs will the indole peaks of PDA at 1625–1640 cm (Fig. 2C). XPS
43
play a decisive role in achieving a higher DT. Therefore, we want analysis confirmed that the PDCBs had a comparative CB weight
to rationally endow the PDCBs with a mild DT and a higher drug- percentage to that in the feed ratio (Fig. 2D). This indicates that
loading capacity. Taking the DT and the drug-loading capacity into the prodrug DA-CB could be rationally copolymerized with DA to
consideration, both the PDCB30 and PDCB40 were used for the produce the polyprodrug type PDCBs. Specifically, the XPS of the
following studies (Table 1 and Table S2, ESI†).
PDCBs shows the binding energy peaks of –S–S– at 163.7 eV,
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 421

View Article Online
Paper
Journal of Materials Chemistry B
–COO– at 533.5 eV, and C–Cl at 201.7 eV, verifying the disulfide the PDCBs in PBS hardly changed for two months at 4 1C
bond, the carboxylic ester linker and the CB drug are retained (Fig. S4C, ESI†). The PDCBs have a negative zeta potential of
in the PDCBs (Fig. S2 and S3, ESI†). As shown in Fig. 3 and about 30.2–32.1 mV, enabling them to remain relatively stable
Fig. S4 (ESI†), both the DLS and TEM analyses show that the in FBS at 37 1C despite the inevitable albumin absorption.
PDCBs have an almost spherical morphology with hydrodynamic
3
.2 Photothermal effect and multi-stimuli-triggered drug
diameters ranging from 120 to 150 nm, and the size determined
using TEM is basically consistent with that measured using DLS.
release of PDCBs
The size and morphology of the control PDA was characterized Generally speaking, native melanin and the PDA-based nano-
using similar methods and is shown in Fig. S5 (ESI†). Taken materials have a NIR absorption of 650–900 nm because of their
together, the PDA-based polyprodrug PDCBs with physiologically hierarchical structures, including the p–p stacking of the indole
related dual stimuli triggers and a higher drug-loading capacity units, charge transfer, and the p-electron radicals localized to
2
1–25
can be successfully prepared in a one-pot synthesis from the the quinone residues.
oxidative precipitation copolymerization of the prodrug DA-CB present a broad and strong NIR-absorption at 650–900 nm in
with DA. addition to a characteristic UV signal for the CB at 225 nm
After the nanoparticles enter into the bloodstream in vivo, (Fig. 4A), and the NIR absorbance showed no observable differ-
The UV-Vis-NIR spectra of the PDCBs
ꢀ
2
the blood proteins nonspecifically absorb onto the surfaces of ence before and after mild NIR irradiation (808 nm, 1 W cm
,
the nanoparticles to form a so-called protein corona, inducing 10 min). These data demonstrate that the PDCBs possess an
aggregation, phagocytosis owing to the mononuclear phagocyte intrinsic photothermal conversion capability and a good photo-
3
5–37
system, and even final clearance.
The PDCBs were dis- thermal stability. Upon mild NIR irradiation, both the PDCB30 and
persed well in PBS (10 mM, pH 7.4) at 4 1C or incubated with PDCB40 solutions were heated by 13.9 1C and 12.8 1C, respectively,
FBS (1%, 5%, and 10%) at 37 1C and their physiological stability in comparison to the control PBS which was only heated by 2.5 1C
was monitored using DLS (Fig. 3B). The PDCBs in FBS (Fig. 4B). Moreover, the heating–cooling cycle curves were repeated
remained stable showing no obvious changes in 8 h and then three times with a negligible DT change (Fig. 4C). This data
increased to o250 nm after 24 h incubation, on the other hand, suggests that the PDCBs have good photostability, which is also
Fig. 3 The dependence of the nanoparticle size on the incubation time in FBS at 37 1C (PDCB30, A; PDCB40, B) and the TEM photographs (PDCB30, C;
PDCB40, D) for the PDCBs. Data are presented as means ꢂ SD (n = 5).
4
22 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
ꢀ
1
ꢀ1
Fig. 4 (A) The UV-Vis-NIR spectra for the PDCBs solutions (PDCB30, 500 mg mL ; PDCB40, 400 mg mL ) or the 60-fold diluted solutions and those
ꢀ2
ꢀ1
under NIR irradiation (808 nm, 1 W cm , 10 min). (B) The temperature-elevating magnitude of the PDA (500 mg mL ) or PDCBs solutions against the
irradiation time. (C) The temperature change curves of the PDCB40 solution during laser on/off cycles. (D) Heating curves of the PDCB40 solution against
the irradiation time with different power intensities.
consistent with the UV-Vis-NIR spectra. Furthermore, the DT and 94%, respectively. After treatment using 10 mM DTT, the
magnitude can be easily tuned using the laser power density maximal UV absorption of PDCB40 red shifted from 225 to
and the nanoparticle concentration (Fig. 4D and Fig. S6, ESI†). 260 nm, which is indicative of the disulfide bond cleavage and
Notably, both the PDCB30 and PDCB40 have a higher photo- production of the CB derivative. These data clearly show that
thermal conversion efficiency of about 37.2% and 39.1%, which the reduction and the combined stimuli of 10 mM DTT + pH 5.0
2
1,24,43
is comparable to 40.3% for pure PDA.
In all, these triggered an accelerated drug release and a synergistic action,
findings demonstrate that the high drug-loading polyprodrug which was caused by the cleavage of the reduction-sensitive
PDCBs intrinsically possesses a good NIR-absorbing cap- disulfide bond and the combined action of the cleavage and the
4
4–46
ability, a high photothermal conversion efficiency, and excellent pH-sensitive hydrolysis (Scheme S1, ESI†).
It should be
photostability.
noted that the DTT trigger mainly produced the adduct of CB
0
Compared to the physically-loaded polymeric nanodrug with 2,2 -dithiodiethanol while both CB and the adduct were
formulations, the polymer–drug conjugate counterparts can found at 10 mM DTT + pH 5.0.
minimize early leaking of the drug into the bloodstream,
enhancing the drug efficacy and reducing the side effects.
During the drug release process, the size change of the
PDCBs under different triggers was monitored on-line using
3
8–42
The next question to answer is upon various triggers to mimic DLS. When the PDCBs were incubated at pH 5.0 or with 10 mM
the intracellular cytosolic reduction (10 mM DTT) and/or acidic DTT at 37 1C, they promptly increased to form larger aggregates
endolysosome milieu (pH = 5), do the PDCBs demonstrate an (about 2–2.5 mm) within about 3 h and then remained stable for
intracellular dual stimuli-sensitive drug release profile? PDCB30 10 h (Fig. 5C and D). This stimuli-sensitive size change of the
and PDCB40 (Fig. 5A and B), released about 41–46% of the drug PDCBs suggests that the hydrophobic CB or its adduct was
at pH 5.0 for 12 h, while only 3–6% of the drug was found at quickly detached from the nanoparticles owing to the acidic
pH 7.4. Moreover, the mass spectroscopy confirmed that the hydrolysis or reduction-triggered cleavage, as monitored using
CB was hydrolyzed at pH 5.0 while no CB was found at pH 7.4 mass spectroscopy (Fig. S7, ESI†). As characterized using
(Fig. S7, ESI†). Upon addition of 10 mM DTT or 10 mM DTT + TEM (Fig. S8, ESI†), bigger aggregates were formed after the
pH 5.0, the accumulative drug release at 37 1C reached 44% PDCBs were incubated at an acidic pH of 5.0 or after a 10 mM
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 423

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 5 The pH- and DTT-cleavable and NIR-triggered drug release profiles (A and B), and the DLS size change (C and D) dependent on the time in PBS
at 37 1C (n = 6).
DTT reduction. This TEM result is basically consistent with the enlarge and balance at about 2.5 h, while the balancing time
results of the DLS. The zeta potential measurements showed reduced to about 2 h and 1.5 h at pH 5.0 + 10 mM DTT and
that the negatively charged nanoparticles of about ꢀ30 mV pH 5.0 + 10 mM DTT + NIR, respectively, further verifying the
transformed into positively charged nanoparticles of 11.2 mV at synergistic actions of these dual or triple triggers. It should be noted
pH 5.0, however, they decreased to ꢀ43.7 mV at 10 mM DTT. that the triple stimuli-triggered drug release for the PDCBs was
Accordingly, this nanoparticle swelling and aggregation was controlled not only by the triggers of pH, DTT, and NIR, but also by
postulated to occur by two stages: (1) the decreased hydro- the diffusion barrier from the nanoparticles, which could induce a
phobic interaction derived from the CB detachment and the delay in the drug release in accordance with the size change of the
3
7,43
electrostatic repulsion derived from the protonated dopamine nanoparticles.
Collectively, the polyprodrug PDCBs present
) at pH 5.0, or the decreased negative charge at intracellular endo-lysosomal pH, cytosolic reduction, and external
0 mM DTT, would swell the original nanoparticles, and (2) the NIR-triggered drug release profiles, which are expected to syner-
hydrogen-bonding interactions derived from the newly produced gistically enhance the CT efficacy.
,4-dihydroxybenzenepropanoic (–COOH) or the thiol-disulfide
bond exchanges might connect the swollen nanoparticles to
+
moiety (–NH
3
1
3
3
.3 In vitro PT-CT, cellular internalization and trafficking,
3
7–39
and in vivo pharmacokinetics
form larger aggregates
(Scheme S1, ESI†). This detailed
ꢀ
2
mechanism needs further investigation in ongoing work. Upon Upon mild NIR irradiation (808 nm, 1 W cm , 10 min) and/or
mild NIR irradiation with a pH 5.0 or 10 mM DTT (i.e., dual incubation with PDA for 48 h, all of the L929, HeLa, and MCF-7
triggers), the accumulative drug release in 12 h increased by cell lines remained alive above 95% (Fig. S9, ESI†). These data
about 39–71% owing to the photothermal effect which accele- show that mild NIR irradiation and PDA itself induce negligible
rated the drug diffusion from the nanoparticles, inducing a fast cytotoxicity in the healthy and carcinoma cell lines. Upon mild
1
0–12
drug release.
Specifically, the triple triggers of pH 5.0 + NIR irradiation and/or with 0.1 mM BSO (as an inhibitor to
DTT + NIR achieved the fastest drug release profile with a 76.4% prohibit the cytosolic synthesis of reductive GSH), both the
drug release in 2 h compared to the dual triggers of pH 5.0 + PDCBS and the free CB were incubated with HeLa and MCF-7
DTT, meaning that the mild NIR irradiation sharply enhanced for 48 h, respectively. The drug-induced cytotoxicity of the
the drug diffusion. Similarly, the dual triggers of pH 5.0 + NIR or PDCBs or free CB is dose-dependent and the cell viability
pH 7.4 + 10 mM DTT + NIR accelerated the nanoparticles to gradually decreased with an increase in the PDCB or the drug
4
24 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
ꢀ
1
ꢀ1
concentration under various conditions (Fig. 6). Upon the 8.45 mg mL (MCF-7) and that of 3.61 mg mL (HeLa) or
ꢀ
1
treatment with 0.1 mM BSO, the IC50 of PDCB40 greatly increased 3.62 mg mL (MCF-7) for the combination PT and the combi-
ꢀ1
from 14.03 to 30.28 mg mL for the HeLa cells and from 9.72 to nation CT, respectively. The combination index (CI) is generally
8.11 mg mL for the MCF-7 cells; similar results were obtained used to evaluate the synergistic effect between different drugs
for PDCB30 when incubated with 0.1 mM BSO. BSO reduced the and/or therapies.
ꢀ
1
2
4
8–51
A CI of less than 1 implies a synergistic
cytosolic GSH concentration, hindering the disulfide bond cleav- effect while a CI greater than 1 or equal to 1 suggest an adverse
age in the PDCBs, thus producing an attenuated drug release or additive action; the smaller the CI, the better the different
4
5
and an inhibition effect.
drugs and/or therapies act synergistically. Both the PDCB40
Does mild photothermia produce a synergistic action on the and PDCB30 gave a low CI of 0.37 (HeLa) or 0.43 (MCF-7) and
polyprodrug release induced CT? Considering the sole photo- 0.41 (HeLa) or 0.49 (MCF-7), respectively. That is to say, upon
thermal action of the PDA component and the drug-loading mild NIR irradiation, PDCB40 gave a better synergistic effect
capacity, the IC50 of the pure PDA nanoparticles (i.e., IC50 (PT)) than PDCB . Thus, the mild PT-CT treatment implemented by
30
was used to represent the single PT of the PDCBs while the IC50 both the PDCBs and the mild NIR irradiation both demonstrate
(
combination PT) was calculated using the PDA percentage an excellent and synergistic antitumor activity for killing HeLa
4
3,48–51
43,48
of the PDCBs for the combination PT.
Based on the and MCF-7 cell lines.
Furthermore, this synergistic effect
PDA concentration, both PDCB40 and PDCB30 gave a IC50 of between the hyperthermia and polyprodrug-induced CT has
ꢀ1
ꢀ1
5
3.76 mg mL (HeLa) or 69.34 mg mL (MCF-7) for a single PT been clarified using the techniques including flow cytometry,
(
Fig. S10, ESI†). Based on the CB concentration, both the fluorescence microscopy, and CLSM.
ꢀ
1
PDCB40 and PDCB gave an IC50 of 14.03 mg mL (HeLa) or
9
To provide further insight into the synergistic effect between
mild PT and CT, we applied flow cytometry and fluorescence
30
ꢀ1
ꢀ1
ꢀ1
.72 mg mL (MCF-7) and 14.15 mg mL (HeLa) or 9.92 mg mL
(
MCF-7) for a single CT, respectively (Fig. 6). As for the combi- microscopy to study the cellular uptake of the PDCBs in MCF-7.
ꢀ1
nation PT-CT, PDCB40 gave an IC50 of 5.23 mg mL (HeLa) or After MCF-7 was incubated with CB, CB + NIR, PDCB40, PDCB40
5
3
+
ꢀ
1
.23 mg mL (MCF-7) for the combination PT and that of NIR, and PDCB40 + BSO at predetermined time intervals, flow
.69 mg mL (HeLa) or 3.49 mg mL (MCF-7) for the combi- cytometry histograms were recorded in Fig. 7 and Fig. S11 (ESI†).
ꢀ1
ꢀ1
ꢀ
1
nation CT; PDCB30 showed an IC50 of 8.44 mg mL (HeLa) or Compared to those obtained without mild NIR irradiation,
Fig. 6 Cytotoxicity of PDCB30 or PDCB40 incubated with HeLa (A and C) or MCF-7 (B and D) for 48 h at different conditions. All cell viabilities were tested
six times (n = 6).
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 425

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 7 Flow cytometry histograms (A, B and C) and the time-dependent fluorescence intensity (D) of MCF-7 incubated with PDCB40 under different
conditions (red, control; purple, 15 min; green, 0.5 h; orange, 1 h; blue, 2 h; and olive, 4 h).
the mean fluorescence intensity of the MCF-7 treated with
PDCB40 + NIR increased by about 1.1-fold at 4 h, which was trafficking would greatly improve the CT.
Efficient cell internalization and subsequent intracellular
35–37
Does photothermia
about 4.8-fold higher than the control treated with CB. These enhance the ability of the nanoparticles to escape the endolyso-
results indicate that the PDCB40 was more efficiently taken up some? After the MCF-7 cells were incubated with PDCB40 (CB
by the MCF-7 compared to CB, and the mild NIR irradiation emits blue fluorescence) for 2 h with/without NIR irradiation,
could further facilitate the cellular uptake owing to the photo- the endolysosomes and the nuclei were counterstained using
4
3,48
thermally enhanced membrane permeability.
This large LysoTracker (green) and PI (red) and observed using CLSM
difference in the cellular uptake for CB and PDCB40 is because (Fig. 8A and B and Fig. S13, ESI†). Without the NIR irradiation,
PDCB40, with a diameter of o200 nm, was probably internalized PDCB40 was observed to mainly enter into endolysosomes owing
by MCF-7 via an energy-mediated endocytosis process compared to the colocalization of the blue and green fluorescence. Upon
to the passive diffusion path adopted for the CB.
mild NIR irradiation, however, a merged blue and red fluores-
As observed using fluorescence microscopy (Fig. S12, ESI†), cence (i.e., a pink color) was clearly observed, which suggests
the MCF-7 cells incubated with PDCB40 showed a time- that more PDCBs escaped from the endolysosome into the
dependent fluorescence intensity upon mild NIR irradiation cytoplasm, along with the many released CB molecules that
5
2–56
or introduction of the BSO inhibitor, which is consistent with entered into the nuclei.
The MCF-7 cells treated with
the above-discussed flow cytometry results. Moreover, the MCF-7 PDCB40 + NIR irradiation showed a lower colocalization value
cells treated with PDCB40 + NIR showed the strongest pink of (16.6 ꢂ 6.5)%, revealing the NIR-triggered higher endolyso-
fluorescence (the merged color of both the blue fluorescence somal disruption and greater escape capability of PDCB40 com-
emitted from the CB and the red fluorescence emitted from pared with that of (72.4 ꢂ 5.7)% without NIR irradiation. These
the PI) in the PI-stained nuclei at 4 h compared to those without results demonstrated that PDCB40 was rapidly trafficked from
NIR irradiation or with BSO or with free CB. This PI-staining the endolysosomes under NIR irradiation owing to the hyper-
observation further supports the above findings, as monitored thermia induced endolysosomal membrane disruption and
5
2–56
using flow cytometry.
instability.
Collectively, the above findings clarified that
4
26 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
Fig. 8 (A) CLSM photographs for the intracellular distribution of PDCB40 incubated with MCF-7 for 2 h with/without NIR irradiation, the scale bar
represents 25 mm. (B) Quantification of the co-localization between PDCB40 and LysoTracker (n = 15). (C) In vivo pharmacokinetics curves for PDCB40
and CB (n = 4).
mild NIR irradiation could facilitate cellular internalization and nanoparticles passively accumulate and penetrate into tumor
the subcellular trafficking of PDCB40, generating a synergistic site, plays a decisive role in addition to the aforementioned
effect between the CT and PT at the cellular level, as verified by long blood circulation of the non-leaking polyprodrug nano-
5
7
MTT, flow cytometry, fluorescence microscopy, and CLSM.
particles. The anticancer drug doxorubicin (DOX) emits a
As PDCB40 has a higher CB-loading, better antitumor activity strong red fluorescence and the physically DOX-loaded nano-
and a smaller CI than PDCB30, the in vivo performance was evalu- particles with minimal drug leakage can be facilely used to
ated using PDCB40 as an example. Following intravenous injec- monitor in vivo tissue biodistribution using a small animal
5
8–61
tions in SD mice and the collection of blood samples at various fluorescent imaging system.
Utilizing strong p–p stacking
predetermined times, the pharmacokinetic profiles of PDCB40 and hydrogen-bonding interactions between the PDA network
ꢀ
1
(
8 mg kg CB equiv.) and the control of the free CB were recorded and the DOX, a certain amount of DOX (52.6 wt%) was physically
in Fig. 8C. As expected, free CB was rapidly cleared from the absorbed onto PDCB40 to form DOX-loaded PDCB40; meanwhile
plasma, whereas PDCB40 achieved a higher bloodstream CB minimal premature DOX leakage was exhibited in PBS at 37 1C,
3
0,62
concentration and a longer elimination half-life time (t1/2b) of inducing a reduced effect on the in vivo biodistribution.
(
Free
1.34 ꢂ 0.17) h, which is about 3.62-fold greater than that found for DOX and DOX-loaded PDCB40 (5.3 mg kg DOX equiv.) were
free CB, (0.37 ꢂ 0.04) h. Moreover, the area under the curve (AUC) intravenously injected via the tail vein into MCF-7 tumor-bearing
ꢀ
1
ꢀ
1
for PDCB40 was 33.12 ꢂ 0.85 mg h L and this is about 33.8-fold nude mice. As shown in Fig. 9A, the fluorescence signal for the
ꢀ1
greater than 0.98 ꢂ 0.02 mg h L found for free CB. In all, the mice treated with DOX-loaded PDCB40 progressively increased in
polyprodrug nanoparticle formulation of PDCB40 obviously pro- the tumor site, and was strongest at 10 h, it then grew weaker,
longs the blood circulation time of CB owing to the dual nanoscale but was still observable at 24 h; on the other hand, the mice
and prodrug attributes, which are expected to be physiologically treated with free DOX quickly attained the strongest signal at 6 h
37–42
stable allowing an efficient CT to be attained in vivo.
and this sharply attenuated and was negligible at 24 h. These
data indicate that the DOX-loaded PDCB40 formulation facili-
tated fluorescent imaging owing to the EPR effect and enabled a
3.4 In vivo biodistribution and photo-mediated imaging
Precise and efficacious cancer therapy using nanoparticles longer blood retention time, while free DOX was quickly cleared
5
8–61
depends on several barrier steps, and the EPR, in which the from the bloodstream.
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 427

View Article Online
Paper
Journal of Materials Chemistry B
Fig. 9 (A) In vivo fluorescent imaging of the tumor-bearing nude mice after intravenous injection of DOX or DOX-loaded PDCB40. (B) Ex vivo imaging of
ꢀ1
the major organs and the tumors. (C) Biodistribution of DOX in different tissues at a dosage of 5.3 mg kg . The data are shown as mean ꢂ SD (n = 3);
(
**) indicates P o 0.01, (***) indicates P o 0.001. (D) Photothermal imaging of the mice upon mild NIR irradiation. (E) The temperature evolution curve of
the tumor site over the irradiation time (n = 4). (F) The PA signal intensity of the tumor over time following intravenous injection of PDA or PDCB40 at
same dose (n = 3). (G) The 3D PA images of the tumor at different time intervals.
To investigate the amount of PDCB40 in the tumor and other DOX-loaded PDCB40, the DOX fluorescence in the tumor site
major organs ex vivo (the heart, liver, spleen, lung, and kidneys), was at its strongest at 10 h compared to the other major organs,
all mice were sacrificed after intravenous injection at different suggesting a selective accumulation of PDCB40 in the tumor
time intervals (2, 6 and 10 h). For the mice treated with (Fig. 9B). Considering the DOX level (% ID/g) in the tumor site,
4
28 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
the mice treated with DOX-loaded PDCB40 showed a 6.5-fold
Next, we evaluated the in vivo photothermal effect and the
greater level DOX at 10 h compared to those used as the control imaging of PDCB40. Three groups of MCF-7 tumor-bearing mice
and treated with free DOX (Fig. 9C). These in vivo and ex vivo were intravenously administrated with PBS, PDA and PDCB40 at
data show that the polyprodrug PDCB40 formulation indeed has a time point of 0 day, respectively. At 10 h after the injection
4
3,58–61
ꢀ2
a selective tumor accumulation effect.
and after mild NIR irradiation (808 nm, 1 W cm , 10 min),
Fig. 10 (A) Photographs of the mice undergoing various treatments. The dependence of the tumor volume (B) or body weight (C) of the mice on time.
D) Survival curve of the mice with various treatments (n = 4); (***) indicates P o 0.001. H&E stained images (E), TUNEL (F), and PCNA (G) of the dissected
tumors after 32 days of treatment and the healed tissue for the PDCB40 + NIR group, the scale bar represents 150 mm.
(
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 429

View Article Online
Paper
Journal of Materials Chemistry B
the tumor-sites of the three groups were monitored on-line survival study, no mice in the PBS, PBS + NIR, PDA groups
using a thermal imaging camera during the irradiation process. remained alive on days 32–36 owing to the oversized tumors
3
As shown in Fig. 9D and E, the tumor-site temperature of the (41500 mm ), whereas none of single CT and single PT groups
mice treated with PDA or PDCB40 increased to about 45 1C from were surviving on day 50 (Fig. 10D). Impressively, all of the mice
34.5 1C after 5 min of NIR irradiation, and then slowly heated in the PT + CT group remained alive with complete and traceless
up to about 47 1C; in contrast, the PBS group only increased by tumor ablation on 50 day after one injection and one mild dose
about 3.8 1C. It should be note that PDCB40 + NIR can elevate of NIR irradiation; however, heavy scar and skin burning
the tumor-site temperature to over 43 1C in 3 min and then kill damage were often observed in the normal PT and PT-CT
the cancer cells from 3 to 10 min as cancer cells are vulnerable groups at more than 50 1C upon a high intensity or high dose
6
3
17–20
to hyperthermia above 43 1C. These data convincingly demon- of NIR irradiation.
strate that PDCB40 can mediate a mild photothermic effect on strate that the mild PT-CT treatment produced a superior and
the tumor upon mild NIR irradiation. synergetic therapeutic efficacy in vivo. In addition, all mice in
These findings convincingly demon-
As an emerging noninvasive imaging modality, photoacoustic the PT-CT group showed little change in their mean weight
PA) imaging is attractive for cancer imaging and diagnosis as during the 50 days compared with the other groups, suggesting
(
both nanoparticles and NIR irradiation result in heat fluctua- minimal systematic toxicity of PDCB40 under mild NIR irradia-
tions and ultrasonic waves, making 2D or 3D imaging possible in tion (Fig. 10C).
3
1,32,43,64
a target containing accumulated nanoparticles.
For the
To further assess the antitumor efficacy of the above treat-
PDA (control) and PDCB40 treated mice (Fig. 9G and F), the PA ments, the dissected tumors and major organs (liver, kidneys,
signal of the tumor site enhanced progressively and reached a spleen, lung and heart) of all of the groups on day 32 were
maximum at 10 h post injection. For the PDCB40 treated mice, stained using H&E for histological analysis. Heavily decreased
the strongest PA signal at 10 h was 4.1-fold higher than that of cellularity, and nuclear shrinkage and fragmentation were
the contrast image at 0 h, although this was a little attenuated observed for the completely-healed tissue (i.e., original tumor
(4.5%) compared with the control PDA; then the PA signal site) in the mild PT-CT group compared to the tumors from the
gradually reduced but remained 1.8-fold stronger than the free CB, mild PT, and CT groups (Fig. 10E). The main organs for
contrast image at 24 h. These findings indicate that PDCB40 all of the mice from the three therapeutic groups (PT, CT, and
has an intrinsic and excellent PA imaging ability in vivo owing to PT + CT) and the three control groups (PBS, PBS + NIR, PDA)
a selective accumulation and retention effect in the tumor site. showed no obvious pathological change and damage, although
It should be noted that the PDA-mediated photoacoustic the kidneys from the mice in the free CB group changed slightly
imaging result was consistent with the above mentioned DOX- (Fig. S15, ESI†). Moreover, both the tumor cell apoptosis and
induced fluorescent imaging results when utilizing the DOX-loaded proliferation were examined using TUNEL and PCNA, respec-
PDCB40. Collectively, the NIR-absorbing PDCB40 intrinsically tively. The mild PT + CT group presented a significant cell
possesses both photothermal and PA imaging modalities, making apoptosis compared to a moderate level in the mild PT, CT and
it useful as a novel nanotheranostic agent for imaging-guided free CB groups, a small amount was observed in the PDA group,
3
5–37
precision cancer therapy.
and a minimal amount was found in the PBS and PBS + NIR
groups (Fig. 10F). On the contrary, the PCNA-positive tumor
cells were reduced the most in the mice from the mild PT + CT
3
.5 In vivo antitumor activity, H&E, TUNEL and PCNA assays
Based on the in vitro and in vivo studies of PDCB , we group, with the next most reduced in the mild PT, CT and free
4
0
comparatively examined the antitumor performance of PDCB40 CB groups, less in the PDA group, and the least in the PBS and
by using MCF-7-bearing nude mice with an average tumor PBS + NIR groups (Fig. 10G). In all, by one intravenous injection
3
volume of 60–80 mm . On day 0, the mice were placed into and one mild dose of NIR irradiation, the polyprodrug formula-
seven groups and intravenously administrated with either PBS, tion of PDCB40 generated a superior and synergistic antitumor
ꢀ1
PBS + NIR, PDA, free CB (8 mg kg ), PDA + NIR (PT), PDCB40 efficacy in vivo and achieved complete and traceless ablation of
(
CT, 8 mg CB equiv. per kg), or PDCB40 + NIR (PT + CT), solid tumors without recurrence in 50 days, providing a mild
4
3,46,63–69
respectively. Note that all the groups were injected once on day PT-CT treatment that is practical for tumor therapy.
0
ꢀ2
with/without one dose of NIR-irradiation (808 nm, 1 W cm
,
10 min). As shown in Fig. 10A, the tumors in mice treated with
the PT-CT group completely regressed without skin scarring or
burning damage on day 6, and did not recur over the 50 day
4. Conclusions
evaluation period. However, the tumors of those from the To address the key dilemmas and enhance the transition
single CT and PT groups increased continuously although their potential of cancer PT and PT-CT treatments, we put forward
tumor growth rates were lower than those in the PBS, PBS + an innovative strategy to leverage the hyperthermia and high
NIR, PDA, and free CB groups (Fig. 10B and Fig. S14, ESI†). drug-loading capacity of a polyprodrug to achieve mild PT-CT
Specifically, compared to the PBS group, the mild PT-CT treat- treatment. We have rationally constructed a novel class of
ment achieved a tumor inhibitory rate (TIR) of 100%, which PDCBs conjugate nanotheranostics, which present good photo-
is significantly higher than that of free CB (35.5 ꢂ 7.3%), thermal properties, have intracellular pH-, reduction-cleavable,
CT (41.1 ꢂ 8.3%), and mild PT (45.4 ꢂ 5.3%). During the and NIR-triggered drug release profiles, and demonstrate a
4
30 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Journal of Materials Chemistry B
Paper
prolonged blood circulation time and selective tumor accumu- 13 S. Zhang, H. Chen, G. Zhang, X. Kong, S. Yin, B. Li and
lation effect. By monitoring the treatments using MTT, flow
L. Wu, J. Mater. Chem. B, 2018, 6, 241–248.
cytometry, fluorescence microscopy, and CLSM, it was shown 14 T. Wang, D. Wang, H. Yu, M. Wang, J. Liu, B. Feng, F. Zhou,
that PDCB40 + mild NIR irradiation could efficiently promote
cellular internalization and subcellular trafficking, producing an
Q. Yin, Z. Zhang, Y. Huang and Y. Li, ACS Nano, 2016, 10,
3496–3508.
excellent and synergistic antitumor effect in vitro. In addition to 15 J. Nam, S. Son, L. J. Ochyl, R. Kuai, A. Schwendeman and
exhibiting good photothermal and photoacoustic imaging cap-
J. J. Moon, Nat. Commun., 2018, 9, 1074.
abilities, the polyprodrug PDCB40 formulation produced superior 16 L. Zhang, Y. Qin, Z. Zhang, F. Fan, C. Huang, L. Lu,
antitumor efficacy using one intravenous injection and a single
mild dose of NIR irradiation, achieving complete and traceless
H. Wang, X. Jin, H. Zhao, D. Kong, C. Wang, H. Sun,
X. Leng and D. Zhu, Acta Biomater., 2018, 75, 371–385.
ablation of a solid MCF-7 tumor without recurrence within 17 X. Zhen, C. Xie and K. Pu, Angew. Chem., Int. Ed., 2018, 130,
0 days. Significantly, this work opens up a new avenue for the
4002–4006.
development of a polyprodrug-based mild PT-CT treatment that 18 L. Pan, J. Liu and J. Shi, ACS Appl. Mater. Interfaces, 2017, 9,
5
allows intrinsic photothermal and photoacoustic imaging for
practical and precise cancer therapy.
15952–15961.
19 Y. Yang, W. Zhu, Z. Dong, Y. Chao, L. Xu, M. Chen and
Z. Liu, Adv. Mater., 2017, 29, 1703588.
2
0 R. Zhu, F. Gao, J.-G. Piao and L. Yang, Biomater. Sci., 2017, 5,
Conflicts of interest
1596–1602.
2
2
1 R. Mrowczynski, ACS Appl. Mater. Interfaces, 2018, 10, 7541–7561.
2 M. Y. Liu, G. J. Zeng, K. Wang, Q. Wan, L. Tao, X. Y. Zhang
and Y. Wei, Nanoscale, 2016, 8, 16819–16840.
There are no conflicts to declare.
Acknowledgements
23 X. P. Chen, Y. S. Huang, G. Yang, J. X. Li, T. T. Wang,
O. H. Schulz and L. K. Jennings, Curr. Pharm. Des., 2015, 21,
C. M. DONG thanks the National Natural Science Foundation
of China (21774074 & 21474061). Financial support from the
Innovation Program of Shanghai Municipal Education Commission
4262–4275.
2
2
4 Y. Liu, K. Ai and L. Lu, Chem. Rev., 2014, 114, 5057–5115.
5 M. D’Ischia, A. Napolitano, V. Ball, C. T. Chen and M. J.
Buehler, Acc. Chem. Res., 2014, 47, 3541–3550.
(201701070002E0061) and the Innovation Fund (IFPM2016B004) of
Shanghai Jiao Tong University & Affiliated Sixth People’s Hospital
South Campus are also appreciated.
2
2
2
6 Y. Li, C. Jiang, D. Zhang, Y. Wang, X. Ren, K. Ai, X. Chen and
L. Lu, Acta Biomater., 2017, 47, 124–134.
7 S. H. Kim, I. In and S. Y. Park, Biomacromolecules, 2017, 18,
1825–1835.
References
8 L. Zhang, H. Su, J. Cai, D. Cheng, Y. Ma, J. Zhang, C. Zhou,
S. Liu, H. Shi, Y. Zhang and C. Zhang, ACS Nano, 2016, 10,
10404–10417.
1
H. S. Jung, P. Verwilst, A. Sharma, J. Shin, J. L. Sessler and
J. S. Kim, Chem. Soc. Rev., 2018, 47, 2280–2297.
2
M. Karimi, P. S. Zangabad, S. Baghaee-Ravari, M. Ghazadeh, 29 X. Wu, L. Zhou, Y. Su and C. M. Dong, J. Mater. Chem. B,
H. Mirshekari and M. R. Hamblin, J. Am. Chem. Soc., 2017,
39, 4584–4610.
R. Batul, T. Tamanna, A. Khaliq and A. Yu, Biomater. Sci.,
017, 5, 1204–1229.
M. Abbas, Q. Zou, S. Li and X. Yan, Adv. Mater., 2017,
9, 1605021.
2016, 4, 2142–2152.
30 X. Wang, J. Zhang, Y. Wang, C. Wang, J. Xiao, Q. Zhang and
Y. Cheng, Biomaterials, 2016, 81, 114–124.
31 J. Lin, M. Wang, H. Hu, X. Y. Yang, B. Wen, Z. T. Wang,
O. Jacobson, J. B. Song, G. F. Zhang, G. Niu, P. Huang and
X. Y. Chen, Adv. Mater., 2016, 28, 3273–3279.
1
3
4
5
6
7
2
2
X. Huang, W. Zhang, G. Guan, G. Song, R. Zou and J. Hu, 32 Q. Fan, K. Cheng, X. Hu, X. Ma, R. Zhang, M. Yang, X. Lu,
Acc. Chem. Res., 2017, 50, 2529–2538.
C. Liang, L. Xu, G. Song and Z. Liu, Chem. Soc. Rev., 2016, 45,
L. Xing, W. Huang, S. S. Gambhir and Z. Cheng, J. Am.
Chem. Soc., 2014, 136, 15185–15194.
6
250–6269.
Y. Chen, L. Wang and J. Shi, Nano Today, 2016, 11,
92–308.
K. K. Ng and G. Zheng, Chem. Rev., 2015, 115, 11012–11042.
33 J. Park, T. F. Brust, H. J. Lee, S. C. Lee, V. J. Watts and Y. Yeo,
ACS Nano, 2014, 8, 3347–3356.
34 M. Zhang, X. Chen, L. Zhang, L. Li, Z.-M. Su and C. Wang,
Chem. Mater., 2018, 30, 3722–3733.
2
8
9
J. B. Song, P. Huang, H. W. Duan and X. Y. Chen, Acc. Chem. 35 J. Shi, P. W. Kantoff, R. Wooster and O. C. Farokhzad, Nat.
Res., 2015, 48, 2506–2515.
Rev. Cancer, 2017, 17, 20–37.
0 L. Cheng, C. Wang, L. Feng, K. Yang and Z. Liu, Chem. Rev., 36 Y. Yan, M. Bj o¨ rnmalm and F. Caruso, ACS Nano, 2017, 3,
014, 114, 10869–10939. 9512–9517.
1 V. Shanmugam, S. Selvakumar and C.-S. Yeh, Chem. Soc. 37 M. Elsabahy, G. S. Heo, S.-M. Lim, G. Sun and K. L. Wooley,
Rev., 2014, 43, 6254–6287. Chem. Rev., 2015, 115, 10967–11011.
2 Z. Zhang, J. Wang and C. Y. Chen, Adv. Mater., 2013, 25, 38 X. Pang, Y. Jiang, Q. Xiao, A. W. Leung, H. Hua and C. Xu,
869–3880. J. Controlled Release, 2016, 222, 116–129.
1
1
1
2
3
This journal is ©The Royal Society of Chemistry 2019
J. Mater. Chem. B, 2019, 7, 415--432 | 431

View Article Online
Paper
Journal of Materials Chemistry B
39 R. Tong, L. Tang, L. Ma, R. Baumgartner and J. Cheng, 55 W. Tao, X. Zhu, X. Yu, X. Zeng, Q. Xiao, X. Zhang, X. Ji,
Chem. Soc. Rev., 2014, 43, 6982–7012.
X. Wang, J. Shi, H. Zhang and L. Mei, Adv. Mater., 2017,
29, 1603276.
56 J. Zhang, X. Zhang, G. Liu, D. Chang, X. Liang, X. Zhu,
W. Tao and L. Mei, Theranostics, 2016, 6, 2099–2113.
4
4
4
0 R. Duncan, J. Controlled Release, 2014, 190, 371–380.
1 J. Kope ˇc ek, Adv. Drug Delivery Rev., 2013, 65, 49–59.
2 Y. Hou, Y. Zhou, H. Wang, R. Wang, J. Yuan, Y. Hu,
K. Sheng, J. Feng, S. Yang and H. Lu, J. Am. Chem. Soc., 57 H. Maeda, H. Nakamura and J. Fang, Adv. Drug Delivery Rev.,
018, 140, 1170–1178. 2013, 65, 71–79.
3 C. Du, J. Qian, L. Zhou, Y. Su, R. Zhang and C. M. Dong, ACS 58 Y. Zhu, J. Zhang, F. Meng, C. Deng, R. Cheng, J. Feijen and
Appl. Mater. Interfaces, 2017, 9, 31576–31588. Z. Zhong, J. Controlled Release, 2016, 233, 29–38.
4 Y. Han, J. Li, M. Zan, S. Luo, Z. Ge and S. Liu, Polym. Chem., 59 D. Li, Z. Tang, Y. Gao, H. Sun and S. Zhou, Adv. Funct.
014, 5, 3707–3718. Mater., 2016, 26, 66–79.
5 S. Lv, Z. Tang, D. Zhang, W. Song, M. Li, J. Lin, H. Liu and 60 L. Wang, Y. Yuan, S. Lin, J. Huang, J. Dai, Q. Jiang, D. Cheng
X. Chen, J. Controlled Release, 2014, 194, 220–227. and X. Shuai, Biomaterials, 2016, 78, 40–49.
6 Y. Zhang, D. Yang, H. Chen, W. Q. Lim, F. S. Z. Phua, G. An, 61 S.-M. Lee, H. Park and K.-H. Yoo, Adv. Mater., 2010, 22,
P. Yang and Y. Zhao, Biomaterials, 2018, 163, 14–24. 4049–4053.
7 Q. Wei, K. Achazi, H. Liebe, A. Schulz, P.-L. Michael Noeske, 62 Y. Gao, X. Wu, L. Zhou, Y. Su and C. M. Dong, Macromol.
2
4
4
4
4
4
2
I. Grunwald and R. Haag, Angew. Chem., Int. Ed., 2014, 53,
1650–11655.
8 X. J. Wu, L. Z. Zhou, Y. Su and C. M. Dong, Biomacromolecules,
016, 17, 489–2501.
9 Q. Y. Hu, W. J. Sun, C. Wang and Z. Gu, Adv. Drug Delivery
Rev., 2016, 98, 19–34.
0 S. Mignani, M. Bryszewska, B. Klajnert-Maculewicz, M. Zablocka
and J.-P. Majoral, Biomacromolecules, 2015, 16, 1–27.
1 L. Ma, M. Kohli and A. Smith, ACS Nano, 2013, 7, 9518–9525.
Rapid Commun., 2015, 36, 916–922.
63 J. O. You, P. Guo and D. T. Auguste, Angew. Chem., Int. Ed.,
2013, 52, 4141–4146.
64 J. Yang, S. Zhai, H. Qin, H. Yan, D. Xing and X. Hu,
Biomaterials, 2018, 176, 1–12.
65 Y. Chen, H. Li, Y. Deng, H. Sun, X. Ke and T. Ci, Acta
Biomater., 2017, 51, 374–392.
1
4
4
5
2
66 Y. Zhang, C. Teh, M. Li, C. Y. Ang, S. Y. Tan, Q. Qu, V. Korzh
and Y. Zhao, Chem. Mater., 2016, 28, 7039–7050.
5
5
5
2 J. Kim, H. Kim and W. J. Kim, Small, 2016, 12, 1184–1192. 67 X. Li, M. Takashima, E. Yuba, A. Harada and K. Kono,
3 T.-Y. Hsieh, W.-C. Huang, Y.-D. Kang, C.-Y. Chu, W.-L. Liao, Biomaterials, 2014, 35, 6576–6584.
Y.-Y. Chen and S.-Y. Chen, Adv. Healthcare Mater., 2016, 5, 68 N. Larson, A. Gormley, N. Frazier and H. Ghandehari,
016–3026. J. Controlled Release, 2013, 170, 41–50.
4 D. Li, M. Zhang, F. Xu, Y. Chen, B. Chen, Y. Chang, 69 C. Q. Luo, Y. X. Zhou, T. J. Zhou, L. Xing, P. F. Cui, M. Sun,
3
5
H. Zhong, H. Jin and Y. Huang, Acta Pharm. Sin. B, 2018,
, 74–84.
L. Jin, N. Lu and H. L. Jiang, J. Controlled Release, 2018, 274,
56–68.
8
4
32 | J. Mater. Chem. B, 2019, 7, 415--432
This journal is ©The Royal Society of Chemistry 2019